Resuscitative Endovascular Balloon Occlusion of the Aorta in a Normovolemic Rabbit Model: a Pilot Study

Background: The purpose of this study was to verify physiological, end-organ and systemic inammatory changes in zone II after resuscitative endovascular balloon occlusion of the aorta (REBOA) in a normovolemic rabbit model. Methods: Anaesthetized rabbits were subjected to aortic balloon occlusion for different times (15 min, 30 min, 60 min and 90 min) followed by 2 h of reperfusion. Rabbits with no balloon occlusion were set as the control group. ELISAs were used to examine the serum levels of ALT, AST, Cr, BUN, MDA, SOD, IL-8, IL-6, and TNF-α; HE staining was used to identify the morphological changes in the kidney; RT-PCR was used to detect the mRNA levels of IL-6, IL-8, TNF-α and NF-κB in the kidney and uterus; and Western blotting was used to measure the protein expression levels of IL-6, IL-8, TNF-α and NF-κB in the kidney and uterus. Results: Plasma concentrations of liver markers, kidney markers, inammatory factors and oxidative stress indicators were signicantly increased at the end of reperfusion in the 30 min, 60 min and 90 min groups. Damage to the kidney occurred in the 30 min, 60 min and 90 min groups. The mRNA and protein expression levels of IL-6, IL-8, TNF-α and NF-κB in the kidney and uterus were signicantly increased at the end of reperfusion in the 30 min group, and as the time of occlusion extended, these levels continued to increase. Conclusion: Activation of systemic inammation and ischaemia-reperfusion injury


Background
Resuscitative endovascular balloon occlusion of the aorta (REBOA) is an effective technology to address noncompressible torso haemorrhage. The procedure involves placing and in ating a balloon catheter in the abdominal aorta under the guidance of imaging equipment. This technique can effectively prevent and reduce fatal bleeding during surgery, ensuring that abdominal surgery, urological surgery, vascular surgery, organ transplantation, tumour resection, obstetrics and gynaecology procedures, and pelvic and orthopaedic surgeries can start in a clear eld [1] .
The aortic occlusion area is often divided into three zones: zone I, comprising the left subclavian artery opening to the celiac artery (upper abdominal aorta); zone II, comprising the celiac artery to the lowest renal artery (middle abdominal aorta); and zone III, comprising the lowest renal artery to the bifurcation of the abdominal aorta (lower abdominal aorta). At present, there are many debates regarding the clinical application of REBOA. Our former clinical studies con rmed that REBOA at the level of the renal artery resulted in less blood loss [2] . The main reasons for balloon occlusion in this area lie in the presence of natural collateral circulations to the uterus from the ovarian arterial blood supply [3] . Kim et al. reported that the frequency of postpartum haemorrhage originating from the ovarian arteries reached 11.8% [4] .
In clinical practice, there is no uniform standard for the duration of blood ow occlusion. A longer period of REBOA will cause ischaemia-reperfusion injury (IRI), which will lead to organ failure. The pathophysiology of IRI is related to microvascular dysfunction, which is characterized by telangiectasia, decreased tissue perfusion, and organ failure [5] . With the extension of REBOA time, the burden of IRI increases, and organ failure occurs [6] . The organs and tissues below the level of occlusion will produce a large number of in ammatory mediators, such as endotoxin, oxygen free radicals, and complement activation of tumour necrosis factor α (TNF-α) and interleukin-1β (IL-1β) due to ischaemia and reperfusion, and other cytokines. These mediators can release large amounts of oxygen free radicals and proteolytic enzymes after entering multiple tissues in the body from the bloodstream, causing tissue damage. In addition, activating cytokines and in ammatory mediators such as interleukin-6 (IL-6), interleukin-8 (IL-8), TNF-α and IL-1β further stimulate and exacerbate the in ammatory cascade. Thus, the REBOA time should strike a reasonable balance between the physiological haemodynamic effects of obstructing blood ow to organs and the time required to effectively control bleeding.
To date, various clinical studies and haemorrhagic animal models have been performed to describe the effects of REBOA in a state of haemorrhagic shock [7][8][9] , but few studies have been conducted to verify the effects of REBOA in normovolemic conditions and provide a thorough description of the changes from 15 min to 90 min of balloon occlusion. Therefore, the purpose of this study was to investigate changes in physiology, IRI and systemic in ammation in response to different times of REBOA in zone II in a non-shock rabbit model.

Animals
The animal instrumentation and ensuing experiments were approved by the institutional review board for animal research. Our study protocol was reviewed and approved by the Institutional Animal Care and Use Committee of Zhengzhou University First A liated Hospital. Forty female New Zealand white rabbits (Hualan Biological Co., Ltd. China) (2.5-3.5 kg) were housed under conditions with a 12 h dark/light cycle and free access to food and water before experimentation.

Animal preparation
Before intervention, the animals were randomized into ve groups: I (control group, no occlusion), II (occlusion for 15 min), III (occlusion for 30 min), IV (occlusion for 60 min), and V (occlusion for 90 min). The hair on the right inguinal region and ventral side of the chest and abdomen was shaved, and the skin was sterilized. Animals were anaesthetized by intravenous (i.v.) pentobarbital sodium 20-30 mg/kg. The core body temperature of rabbits was kept at 37°C throughout experiments with heating pads. The zone II aorta was occluded with a balloon catheter for different times and then reperfused for 2 h. Then, all animals were sacri ced with air injection via the ear marginal vein before the collection of sera, renal tissues and uterine tissues.

Experimental protocol
A 2F Fogarty balloon catheter (Guorui Jiachuang Technology Co., Ltd. China) was inserted through the femoral artery sheath, and the balloon position was adjusted with digital subtraction angiography (DSA) ensuring that the upper edge of the balloon was placed against the opening of the higher renal artery. Indirect abdominal aortic angiography was performed through the right ear vein to con rm the effect of occlusion. After the occlusion time was reached, the balloon was slowly de ated immediately to allow 2 h of blood reperfusion, at which time the rabbits were sacri ced to collect serum, renal tissue and uterine tissue.

Evaluation of hepatic and renal function and in ammatory factors
Whole blood from rabbits was centrifuged for 10 min at 3000 rpm/min, and then the serum layer was collected for the detection of the levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), serum creatinine (Scr) and blood urea nitrogen (BUN), which were examined by the laboratory in our hospital. The levels of MDA, SOD, IL-8, IL-6, and TNF-α in the serum were detected with ELISA kits.

Histopathology
The renal samples were xed in 4% paraformaldehyde for 12 h followed by para n embedding, sectioning and staining. Histomorphological evaluation was performed by the pathologist using a double-blind method.
Determination of the mRNA levels of IL-6, IL-8, TNF-α and NF-κB in the kidney and uterus Total RNA was extracted using TRIzol (Aidlab) following the manufacturer's instructions. cDNA was synthesized with the HiScript Reverse Transcription kit (VAZYME). The primers used are shown in Table 1.
Reverse transcription for RT-PCR was carried out with the following program: 95°C for 10 min followed by 40 cycles of 95°C for 30 s, 50°C for 2 min and 60°C for 30 s. The data were calculated by the 2-ΔΔCt method as described in the instructions.

Statistical analysis
Statistical analysis was performed with SPSS 19.0 statistical software. Measurement data are expressed as the means ± standard deviation (x±s). Comparisons among three groups or more were performed by one-way ANOVA. The test level was α = 0.05, and P <0.05 was considered statistically signi cant.

Animal models
All experimental rabbits were successfully modelled. Delayed angiography under DSA showed that blood ow in the lower abdominal aorta below the level of balloon occlusion was completely interrupted, as there was no contrast agent visible below the balloon catheter ( Figure 1).

Liver and kidney function indicators
After the aortas were occluded for the indicated times in each group, the BUN levels in groups IV and V were 7.93 ± 1.12 mmol/L and 9.67 ± 1.63 mmol/L, respectively, which were signi cantly higher than those in group I (4.93±0.39 mmol/L, P<0.05 for both). The Scr levels in groups IV and V were signi cantly higher than those in group I (58.86 ± 8.21 and 68.71 ± 5.82 mmol/L vs 50.0±3.91 mmol/L; P<0.05 for both). AST levels in groups IV and V increased signi cantly to 60.8±9.44 U/L and 76.85 ± 6.59 U/L, respectively, compared with that in group I (21.28 ± 4.03 U/L; P<0.01 for both). The ALT levels in groups III, IV and V were 29.0 ± 1.82 U/L, 52.14 ± 4.37 U/L, and 80.28 ± 4.99 U/L, respectively, which were signi cantly higher than that in group I (24.71±4.27 U/L; P <0.05 for all; Figure 2).

In ammatory factors
After 2 h of reperfusion, the IL-6, IL-8, and TNF-α levels in groups III, IV and V were all increased compared with those in group I. The concentrations of IL-6 in groups III, IV and V were 24.6±1.73 pg/mL, 34.41±3.33 pg/mL and 42.58±3.14 pg/mL, compared with 15.41±1.0 pg/mL for group I (P <0.01 for all); the IL-8 concentrations in groups III, IV, and V were 172.15±20.65 pg/mL, 209.15±27.3 pg/mL, 289.03 ± 18.69 pg/mL, respectively, compared with 85.62±9.66 pg/mL for group I (P<0.01 for all); and the TNF-α concentrations in groups III, IV and V were 182.28±17.72 pg/mL, 232.43±19.32 pg/mL and 323.86±13.47 pg/mL, respectively, compared with 127.14±10.23 pg/mL for group I (P<0.01 for all; Figure 3).

Oxidative stress indicators among groups
The MDA levels in groups III, IV and V increased to 3.17±0.28 nmol/mL, 3.59±0.23 nmol/mL and 4.91±0.28 nmol/mL, respectively, compared with the level in group I (1.61±0.12 nmol/mL; P <0.01 for all); however, the SOD levels in groups III, IV and V gradually decreased to 138.0±13.21U/mL, 112.71±7.47U/mL and 87.0±9.63U/mL, in comparison with the level in group I (163±6.74 U/mL), and the difference was statistically signi cant (P <0.01 for all; Figure 4).

The pathology of kidney tissue
The kidney tissue structure in groups I and II was clear, with the structure of the glomerular cells and renal tubular cells showing mainly normal features. In group III, mild swelling of glomerular cells and diffuse swelling of renal tubular cells were observed. In group IV, severe diffuse swelling and some necrosis of the renal tubular cells was observed, and moderate to severe swelling of glomerular cells was seen. In group V, the degree of swelling of the glomerular cells and renal tubular cells was more severe, and haemorrhage was present in the renal interstitium ( Figure 5). 6 The expression levels of IL-6, IL-8, TNF-α and NF-κB mRNA in kidney and uterine tissues 6.1 Kidney tissues: IL-6 mRNA levels in groups III, IV and V were 1.267±0.052, 2.221±0.258 and 2.749±0.142, respectively, compared with those in group I (0.884±0.026), the differences were statistically signi cant (P<0.01 for all). IL-8 mRNA levels in groups III, IV and V were higher than those in group 1 (1.402±0.073, 2.056±0.241 and 2.269±0.374, respectively, vs. 0.857±0.126), and the differences were statistically signi cant (P<0.05, all). TNF-α mRNA levels in groups III, IV and V were 1.486±0.105, 1.987±0.211 and 2.954±0.176, respectively, which were higher than those in group I (0.823±0.089), and the differences were statistically signi cant (P<0.05, all). Finally, the NF-κB mRNA levels in groups III, IV and V were 1.636±0.279, 2.330±0.643 and 3.243±0.274, which were higher than those in group I (0.958±0.069), the differences were statistically signi cant (P<0.05 for all). The data are summarized in Figure 6.
6.2 Uterine tissues: The IL-6 mRNA levels in groups III, IV and V were 1.278±0.088, 1.635±0.174 and 2.028±0.091, respectively, which were higher than those in group I (0.947±0.084), and the differences were statistically signi cant (P<0.05 for all). IL-8 mRNA levels in groups III, IV and V were 1.357±0.159, 1.852±0.121 and 2.232±0.103, respectively, which were higher than those in group I (0.979±0.041), and the differences were statistically signi cant (P<0.05, all). TNF-α mRNA levels in groups III, IV and V were 1.470±0.208, 1.626±0.242 and 2.113±0.069, respectively, which were higher than those in group I (0.987±0.062), and the differences were statistically signi cant (P<0.05, all). Finally, NF-κB mRNA levels in groups III, IV and V were 1.770±0.118, 2.105±0.097 and 2.542±0.098, respectively, which were higher than those in group I (0.944±0.073), and the differences were statistically signi cant (P<0.05 for all). The data are summarized in Figure 7.
7 The protein expression levels of IL-6, IL-8, TNF-α and NF-κB in the kidney and uterine tissues 7.1 Kidney tissues: IL-6 protein expression in groups III, IV and V was 0.446±0.064, 0.592±0.046 and 0.673±0.062, respectively, which was higher than that in group I (0.317±0.024), and the differences were statistically signi cant (P<0.05 for all). IL-8 protein expression in groups III, IV and V was 0.427±0.007, 0.581±0.021 and 0.640±0.165, respectively, which was higher than that in group I (0.301±0.029), and the differences were statistically signi cant (P<0.05 for all). TNF-α protein expression in groups III, IV and V was 0.583±0.090, 0.738±0.070 and 0.844±0.086, respectively, which was higher than that in group I (0.342±0.067), and the differences were statistically signi cant (P<0.05 for all). Finally, NF-κB protein expression in groups III, IV and V was 0.556±0.051, 0.713±0.068 and 0.802±0.079, respectively, which was higher than that in group I (0.314±0.039), and the differences were statistically signi cant (P<0.05 for all). The data are summarized in Figure 8. 7.2 Uterine tissues: IL-6 protein expression in groups III, IV and V was 0.284±0.425, 0.426±0.055 and 0.487±0.091, respectively, which was higher than that in group I 0.154±0.034, and the differences were statistically signi cant (P<0.05 for all); IL-8 protein expression in groups III, IV and V was 0.179±0.033, 0.286±0.013 and 0.349±0.032, respectively, which was higher than that in group I (0.118±0.016), and the differences were statistically signi cant (P<0.05 for all). TNF-α protein expression in groups III, IV and V was 0.537±0.025, 0.661±0.038 and 0.773±0.024, respectively, which was higher than that in group I (0.329±0.001), and the differences were statistically signi cant (P<0.05 for all). Finally, NF-κB protein expression in groups III, IV and V were 0.666±0.033, 0.892±0.041 and 0.979±0.039, respectively, which was higher than that in group I (0.454±0.009), and the differences were statistically signi cant (P<0.05 for all). The data are summarized in Figure 9.

Discussion
Until now, there has been no general consensus regarding the optimal occlusion time of the upper and middle abdominal aorta. Ciscato JG et al used vascular clips to clamp the abdominal aorta above the trunk of the celiac artery in 27 dogs and found that 60 min of occlusion followed by 30 min of reperfusion did not cause IRI of kidney tissue [10] . Other scholars, such as Chao CS, found that when the occlusion time of the zone II abdominal aorta in pigs exceeded 30 minutes with a balloon catheter, the intra-abdominal pressure increased, pathological changes in multiple organs occurred, and the mixed venous oxygen saturation was reduced [11] . Athanasiadis D et al concluded that Scr levels increased 30 min after abdominal aorta occlusion above the celiac trunk in pigs with open aneurysms [12] .
The results of this study showed that the BUN and Scr of rabbits were increased after ischaemia for 60 min followed by reperfusion for 2 h, and ALT increased signi cantly after ischaemia for 30 min followed by reperfusion for 2 h. The pathological results of the kidney in this experiment con rmed that swelling of glomerular cells and diffuse swelling of renal tubular cells appeared when the occlusion time reached more than 30 min. The degree of damage increased as the occlusion time increased. However, no IRI of pathological tissues appeared after 30 min of occlusion followed by reperfusion for 2 h. Therefore, it could be concluded that 30 min of occlusion was not safe because IRI occurred. This is basically consistent with the view of Yasuhito et al, who blocked the renal artery to perform a kidneysparing renal tumour resection, and diffuse and irreversible damage of the kidney occurred when the ischaemia time reached 25 min [13] . IRI occurs when tissues or organs are subjected to ischemia for a certain period of time followed by restoration of blood ow or oxygen supply. The pathophysiological changes mainly included the production of oxygen free radicals, in ammatory factors, and lipid peroxidation metabolites. SOD, an important antioxidant enzyme in the body, is a metalloproteinase that can protect cells from damage due to oxygen free radicals. As a product of lipid peroxidation, MDA re ects the level of oxidative stress in the body. The results of our experiment showed that when the occlusion time was longer than 30 min, injuries occurred after reperfusion for 2 h, which were re ected by a signi cant increase in MDA levels and a severe decrease in SOD activity. Aslan M et al. also reported that when an IR model was established in mice, MDA levels increased and SOD activity decreased signi cantly in the uterus and ovaries [14] .
When ischaemia-reperfusion was achieved, tissues were more vulnerable to in ammation. Cytokines, chemokines and neutrophils play an important role in this process. A large number of in ammatory mediators and cytokines, such as IL-1, IL-6 and TNF-α, are released after activation of neutrophils and endothelial cells [15] . In this experiment, the mRNA expression and protein levels of IL-6, IL-8, TNF-α and NF-κB increased as the occlusion time increased extended. After more than 30 min of occlusion, signi cant expression differences in the kidneys, uterus and ovaries were observed in those values after reperfusion for 2 h. This study is consistent with the results reported by Nayki C and others that IL-1β and TNF-α levels increased signi cantly after ovarian ischaemia-reperfusion [16] . Many studies have shown that when ischaemia-reperfusion occurs, the levels of IL-6, IL-8 and TNF-α are all signi cantly increased [17,18] . NF-κB can induce the gene expression of a variety of cytokines, adhesion molecules and chemokines; regulate in ammation upstream; and promote the occurrence of I/R injury [15,19] . NF-κB can regulate the production of cytokines, including TNF-α, IL-1β and IL-6, at the transcriptional level [20] . The results of this experiment showed that when the abdominal aorta was blocked for more than 30 min followed by 2 h of reperfusion, the mRNA and protein levels of NF-κB in the distal tissues increased signi cantly. Liu G et al con rmed in mice that when myocardial ischaemia-reperfusion occurs, the expression level of NF-κB increases and regulates the expression of a variety of in ammatory factors, including increases in IL-1β and TNF-α expression [21] .
There are some limitations to our experimental model. First, the short reperfusion time is not long enough to allow investigation of long-term damage to organs. Second, the number of animals used was small, and further study with a large cohort of rabbits may be needed.
A longer period of occlusion may cause irreversible damage to the body, and the risk of organ necrosis simultaneously increases. When the occlusion is removed, the metabolites accumulated in the ischaemic tissues immediately enter the systemic circulation, which will further aggravate the damage to the organs. Therefore, it is necessary to strictly control the occlusion time to avoid serious damage to the body's multiple organ functions.

Conclusions
As long as the occlusion time was limited to 15 minutes, no signi cant IRI the distal tissues and organs occurred. Therefore, 15 minutes should be regarded as the safe occlusion time.

Declarations
Ethics approval and consent to participate The animal instrumentation and ensuing experiments were approved by the institutional review board for animal research. Our study protocol was reviewed and approved by the Institutional Animal Care and Use Committee of Zhengzhou University First A liated Hospital Consent for publication

Not applicable
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request

Competing interests
The authors have no con icts of interest to declare.

Funding
There is no funding related to this study.
Authors' contributions Juanfang Liu was a major contributor in writing the manuscript and a designer in experiment. and interpreted the animal data. Jianhao Zhang performed the interventional procedure of rabbits. Shanshan Xie collected and interpreted the animal data. Yingxia Liu performed the histological examination of the kidney. Xueliang Zhou did the bio-experiment. Zhaonan Li analyzed the animal data. Xinwei Han organized the whole experiment and rewrote the manuscript. All authors read and approved the nal manuscript.  Figure 2 Hepatic and renal function among groups before occlusion and after 2 h of reperfusion * P<0.05